Saturday, August 18, 2007

Normative Reasons and Motivation

Today, I want to address ADHR's comments:

It's not incoherent to insist that some x is intrinsically valuable (people should want to pursue or promote x) and yet people may not want to pursue or promote x. The simplest case of this is when people fail to recognize that x is intrinsically valuable; more complex cases may involve some sort of akrasia (weakness of the will).

I think this line of argument would have more punch if you turned it squarely against the internalism, arguing that there's something strange about the idea that normative reasons must motivate us to act.

Taking the first item first.

There is a distinction between two different families of claims that is important here.

There is the question of whether it is coherent to say that X is intrinsically valuable, but A is not motivated to bring about or preserve X. ADHR says that it is coherent, as when X is intrinsically valuable but A is not aware of this value.

I agree with this.

ADHR brings up another example of 'akrasia' or 'weakness of will'. This is when a person realizes that something has intrinsic merit, but is not sufficiently motivated by this intrinsic value to overcome basic desires. For example, an alcoholic may realize the virtue of sobriety but not sufficiently to give up the habit of drinking.

In light of this, ADHR thinks that I should argue that there is something strange about the idea that normative reasons must motivate us to act.

As if I agree with him on this statement.

Which, I do.

Agree with him, that is.

Sort of.

Okay, let's look at the details.

All 'ought' or 'should' statements make some type of reference to reasons for action. To say that something 'ought' or 'should' be done, while asserting that there is no reason for doing it, is more than strange, it is incoherent.

I hold that the only reasons for action that exist are desires.

In recent discussion we have distinguished between drive-desires and value-desires. On this distinction, the only reasons that exist are drive-desires. What passes for 'value desires' is simply the recognition that some drive-desires are more useful than others, when it comes to fulfilling other drive-desires. Thus, our drive-desires give us reason to promote some drive-desires over others, and even reason to inhibit drive-desires that thwart other desires.

So, desires are the only reasons-for-action that exist. People still believe in other types of reasons for action; intrinsic value, divine essence, whatever. People can make sensible 'ought' or 'should' statements by referring to these reasons for action. However, because these reasons for action do not exist, any argument built on these types of reasons for action is built on false premises. Their conclusion might still be true, but it is only accidentally or coincidentally true.

So, if every true normative claim talks about reasons for action, then does it not follow that normative reasons must motivate us to act? What sense does it make to say that I have a reason for action that I can sensibly ignore and refuse to act on – or refuse to consider, even if my considerations are outweighed by even stronger reasons not to act.

Because . . .

Okay, desires are the only reasons for action that exist. However, the only desires that motivate my actions are my desires. If you want to motivate me to act, then you need to show me how a state of affairs relates to my desires. Telling me how a state of affairs relates to desires that are not mine – to my neighbor’s desires, for example – will motivate me to act only insofar that I am motivated to see my neighbor’s desires fulfilled. That is to say, I must have a desire that my neighbor’s desires are fulfilled, or at least a desire that is fulfilled by a state in which my neighbor’s desires are fulfilled. If this is not the case, then I have no reason to act.

If, as it turns out, I hate my neighbor, then your news that X will fulfill my neighbor’s desires might motivate me to act so as to make X impossible. This way, I can thwart my neighbor’s desires (which I may well have motivation to do based on my own desires).

The thesis that ‘normative reasons must always motivate us to act’ turns out, in the real world, to mean that ‘the only normative reasons that exist, when it comes to my actions, are my own desires’.

I have argued earlier that my actions are only truly mine if they come from my desires. If Tim over there, with his remote control, is able to move my body, rather than me, then the fact that the actions are proximately connected to his desires and not mine means that the actions being performed are his actions and not mine. If he should direct this body to kill his ex-girlfriend, he would be the one guilty of murder, but not me. He might be able to engineer that I be convicted of that murder, but I (meaning this bundle of desires that I have) am not morally responsible for that murder. He is – because his desires were proximately at the helm of those actions.

So, if normative reasons must motivate us to act, then the only normative reasons that exist for or against my actions are my desires – in precisely the strength that I desire them. Other people’s desires are relevant only insofar as other people serve as a means to the fulfillment of my own desires.

Objectively, relationships between states of affairs and desires that are not mine are just as real as relationships between states of affairs and desires that are mine. Not only are they real, but they are important. We have a lot of very good reasons to talk about and to think about relationships between states of affairs and desires that are not our own.

These desires that exist that are not my desires are still reasons for action, even if they are not reasons for my action. They still motivate people to act in different ways, even though they do not motivate me to act in those ways. They are reasons that people can talk about that are not their own and may not motivate them, but reasons that are real and that a rational person would be unwise to ignore.

As it turns out, other people’s desires are reasons for action for me in a sense in that it is reasons for action for them to do things that affect me. So, if people generally have a reason for action to promote X, then they have a reason for action to change my desires, or to create an environment where I can best fulfill my desires by also promoting X. It means that they have reason to thwart my desires if my desires would get in the way of promoting X. All of these relationships connect the reasons for actions of others with my reasons for action. However, none of them, by their mere existence, necessarily motivate me in any way.

Why would we have invented a language where people were only allowed to talk about relationships between states of affairs and their own desires, but not allowed to talk about relationships between states of affairs and desires not their own? If we are permitted to talk about relationships between states of affairs and desires not our own, what language would we use to do this?

Normative language discusses all reasons for action; mine, yours, ours, theirs, even reasons for action that do not exist. Or, if it isn’t, then we are in desperate need to make some significant adjustments to our language.

Specifically, I argue that moral claims have to do with what reasons for action people generally have reasons to act so as to promote and inhibit. If you were to tell me that ‘people generally have reason to act so as to promote a love of truth,’ this does not imply that I have a reason to act so as to promote love of truth (though, as a part of ‘people generally’, it is more likely that I also have reason to promote a love of truth than that I do not).

5 comments:

Miguel Picanco said...

Thanks for clearing all that up.. I can't wait for you to move on to a new, unrelated topic!

Richard said...

A few objections spring to mind:

(1) Drives aren't reasons. Consider the reluctant drug addict, compelled (against his will) to get his next fix. How is this any different from a mad scientist controlling him via remote control? We may be coerced from within, just as from without. Merely coming from a drive in your head does not suffice to make an action 'yours'.

Any minimally adequate action theory must account for the difference between intentional ("free" - though of course not contracausal) action and mere behaviour. This requires differentiating value-desires (goals) from mere drives, and giving pride of place to the former.

(2) Weighting Desires. What dimension of 'strength' are you referring to? Felt strength? Behavioural impact (motivational force)? Degree of reflective endorsement? These are all logically distinct, though you seem to be conflating them.

Insofar as you treat desires as being revealed by behaviour, you seem to be assuming the second option. But this seems to be the least normatively relevant. (Why should a mere 'cause of behaviour' be thought to reveal what's worth doing?) I care about the goals I endorse, not necessary just whatever moves my body. (Fortunately for me, these usually coincide! But again, there's no reason in principle why this must be so.)

(3) Begging the question on motivation - you write: "The thesis that ‘normative reasons must always motivate us to act’ turns out, in the real world, to mean that ‘the only normative reasons that exist, when it comes to my actions, are my own desires’."

This is simply to assume what is in question, namely: whether desires/motivation can follow from evaluative beliefs (judgments about what we have normative reason to do), or if desires are only ever an input to practical reasoning, and never an output.

(4) Whose reasons? I recall we argued about this last year some time, but "I have no reason to phi" just means "there is no reason for me to phi", which in turn entails that "it is not the case that I ought to phi." So if you have no normative reason to promote truth, then it is (by definition) false that you ought to. Similarly for morality itself.

You write: "other people’s desires are reasons for action for me in a sense in that it is reasons for action for them to do things that affect me."

But that is not any "sense" at all. For other people to have reason to manipulate you into phi-ing, is not in any coherent sense the same thing as you having normative reason to phi. (It wouldn't even be the same thing if the manipulator in question were yourself! See my discussion of acting upon yourself.)

ADHR said...

Alonzo,

I'm having trouble following this.

You go from this: "To say that something 'ought' or 'should' be done, while asserting that there is no reason for doing it, is more than strange, it is incoherent." to this: I hold that the only reasons for action that exist are desires. But the sense of "reason" in the former is not the same -- or, at least, need not be the same -- as the sense of "reason" in the latter. (FWIW, denying this doesn't even require denying that desires exist, or that they are reasons.) In the latter case, we speak of a particular person's reasons. Desires don't float free, after all; so, what the latter claim amounts to is: "The only reasons a particular person can have for action are (their) desires." In the former claim, though, the "reason" in question is a reason for any person. It is general, not particular. So, the former becomes: "To say that something 'ought' or 'should' be done, while asserting that there is no reason for anyone to do it, is incoherent."

Given that, it's obvious that no sort of motivational internalism about normative reasons follows. Normative reasons are reasons that should have motivational purchase on someone's action; motivating reasons are the reasons that actually do have purchase on a particular person's action. So, your claim that "the only normative reasons that exist, when it comes to my actions, are my own desires" doesn't follow. Normative reasons concern the desires I should have, and that someone must have; motivating reasons concern the desires I do have, which may not be the ones I should have.

From what you say in the last two paragraphs, though, I think you accept this sort of distinction. But I have trouble squaring it with the comments I've drawn attention to above.

I also have trouble squaring it with the intersubjectivity you appeal to as constituting normative reasons. I think the better way to characterize the distinction is to appeal to an idealized rationality. Normative reasons are the desires I would have if I were perfectly rational; since I am not perfectly rational, there is a gap between my normative reasons and my motivating reasons. This means that the desires I actually have are my motivating reasons, but the idealized desires are my normative reasons. (There's a question here about whether everyone would have the same normative reasons, but let's leave that aside for now.)

Alonzo Fyfe said...

ADHR

You go from this: "To say that something 'ought' or 'should' be done, while asserting that there is no reason for doing it, is more than strange, it is incoherent." to this: I hold that the only reasons for action that exist are desires. But the sense of "reason" in the former is not the same.

Well, they are not logically the same, but they are materially the same. If I were to say that the only life that exists is on Earth, I am not denying the possibility of life elsewhere, though I would be denying the actual fact of life elsewhere. When I say that the only reasons for action that exist are desires, I am not denying the possibility of other types of reasons for action. However, I am denying that other types of reasons for action exist in fact.

In the former claim, though, the "reason" in question is a reason for any person.

This is ambiguous. It could mean, "reason for all people" - which might exist, but is extremely unlikely and there is no award for special merit even if a desire turns out to be universal.

The other option means, "Considering all reasons regardless of whose they are." Ultimately, these are the reasons that I would argue are a part of moral calculations. Yet, this comes up against the problem that I have no necessary reason to be motivated by your desires (and vica versa). So, many of the reasons involved here are reasons that only contingently motivate an agent, need not motivate him at all, and could even motivate him to do the opposite.

The first option is 'intersubjectivity' (which I reject).

The second option is utilitarian, which I accept.


Given that, it's obvious that no sort of motivational internalism about normative reasons follows.

Yes, on the broad definition of normative reasons (that include all desires). It is not true on the narrow definition of normative reasons (that includes only the desires of the agent.)

Please note, my statement was meant as a reductio of the claim that normative reasons must motivate. Because, if normative reasons must motivate, then only the desires of the agent are normative reasons. However it is not the case that only the desires of the agent are normative reasons. Therefore, it is not the case that normative reasons must motivate.

Normative reasons are the desires I would have if I were perfectly rational; since I am not perfectly rational, there is a gap between my normative reasons and my motivating reasons.

There is no rationality of desires. We can speak about the desires I would have if they were in harmony - and there is a type of means-ends rationality involved in choosing a harmony of desires - but it is a rationality of means, not a rationality of ends. (Though, it is a rationality of means applied to ends.)

If I have a desire that A, B, and C, of equal strength, and A implies not-B, and C implies not-B, then I have reason to exchange desire B for desire not-B. This is because Desire B is in conflict with two equally strong desires. On the other hand, if A implies B, and C implies B, then I should choose desire B rather than not-B, because not-B would generate conflict.

This is the 'rationality of desires' and it has to do exclusively with the relationships that a desire has to other desires. The above analysis goes through regardless of the content or object of these three desires. The object is irrelevant, except insofar as it determines a level of conflict with other desires.

Richard said...

Let me quickly add: if you don't have time to address all my objections, I would wish to highlight the first two as being especially pressing.